Body Mass Index Reconsidered

We know (or should know) just how subjective body image can be, and the psychic toll it takes on millions.


Advertising promotes a generalized message that there is something wrong with the way all of us look, with weight factoring in with all sorts of other things like complexion, age, hair, and height–as it zeros in on particular parts of us that need fixing. Today’s Wall Street Journal (of all places) carried an essay on just how wrong the BMI can be, excerpted briefly below:

“Some researchers say that while BMI improved on its predecessors, it fails to distinguish between different kinds of body mass and therefore can mislead about individuals’ health levels — a longstanding criticism of the measure that hasn’t prevented it from becoming the primary tool for grouping people into normal-weight, overweight and obese categories.

“’When it’s applied to single individuals to make a statement [about their health], it’s nonsense,’ said Stanford University Mathematician Keith Devlin. He has been rallying against BMI, and its classification of his competitive-cycling-shaped figure as overweight, for several years. He trained for a 130-mile, mountainous one-day race in 2009, and in the process added enough muscle mass to cross the overweight threshold, and was stunned to be counseled on weight loss by his doctor.

“’To present it as number with an aura of science and mathematics, that, I think is dishonest,’ Devlin said. ‘I don’t like my discipline being misused.’

“Steven Heymsfield, executive director of the Pennington Biomedical Research Center, a National Institutes of Health-funded institution in Baton Rouge, La., said of BMI, ‘It is reliable as a measure of over-weight but not over-fat. There are both false positives and false negatives, people who are overweight but not over-fat and vice versa.’ False positives are ‘not uncommon in certain populations, for example the military.’

“Heymsfield estimates such false readings have a prevalence of 5% or 10% of the population, but it could be higher. It is possible, for instance, that enough healthy, muscular people are in the overweight category, and enough sick people who have lost weight from illness are in the normal-weight range, ‘that you could completely explain away these results,’ said Heymsfield of results from a recent Journal of the American Medical Association-published study suggesting that people who are moderately overweight have less risk of dying in a given period than people of normal weight. ‘We don’t really know the true proportion of people falsely classified.”


For more, see:

Leave a Reply

Your email address will not be published. Required fields are marked *